ABSTRACT
Hydrolysis of the amino groups in condensed 2,4-diaminopyrimidine systems (1) has been used as a common method for the synthesis of oxo-substituted pyrimidines. In particular, the treatment with 6 M HCl usually yields exclusively the 2-amino-4-oxopyrimidine isomer (2). During our work, we found that the hydrolysis of the amino groups present in some condensed 2,4-diaminopyrimidine systems unexpectedly afforded exclusively the 4-amino-2-oxopyrimidine isomer (3). In this paper, we present the experimental work and ab initio calculations carried out to understand this discrepancy. As a part of such study, eight compounds containing a 2,4-diaminopyrimidine moiety were calculated in gas phase and in aqueous solution, and some acid hydrolyses were reexamined. Results showed that the presence of an electron-donating nitrogen linked to C6 of the 2,4-diaminopyrimidine ring changes the preferred hydrolysis site to yield the 4-amino-2-oxopyrimidine isomer.
Subject(s)
Folic Acid Antagonists/chemistry , Pyrimidines/chemistry , Tetrahydrofolate Dehydrogenase , Chemical Phenomena , Chemistry, Physical , Folic Acid Antagonists/chemical synthesis , Hydrolysis , Indicators and Reagents , Pyrimidines/chemical synthesisABSTRACT
We recently described the syntheses of 12a-c, 4-amino-7-oxo substituted analogues of 5-deaza-5,6,7,8-tetrahydrofolic acid (5-DATHF), and 5,10-dideaza-5,6,7,8-tetrahydrofolic acid (DDATHF), in six steps from commercially available p-substituted methyl benzoates in 20-27% overall yields. Such analogues were tested in vitro against CCRF-CEM leukemia cells and showed that they are completely devoid of any activity, the IC(50) being higher than 20 microg/mL for all cases. To clarify if the presence of the carbonyl group in position C7, the distinctive feature of our synthetic methodology, is the reason for this lack of activity, we have now obtained the 7-oxo substituted analogues of 5-DATHF and DDATHF, 18a-c, in 10-30% overall yield. Testing of 18a-c in vitro against CCRF-CEM leukemia cells revealed that these compounds are totally inactive. A molecular modeling study of 18b inside the active site of the complex E. coliGARTFase-5-DATHF-GAR pointed to an electronic repulsion between the atoms of the 7-oxo group and the carbonyl group of Arg90 as a possible explanation for the inactivity of 18a-c.
Subject(s)
Antineoplastic Agents/chemical synthesis , Tetrahydrofolates/chemical synthesis , Antineoplastic Agents/chemistry , Antineoplastic Agents/pharmacology , Crystallography, X-Ray , Drug Screening Assays, Antitumor , Escherichia coli/chemistry , Folic Acid Antagonists/chemical synthesis , Folic Acid Antagonists/chemistry , Folic Acid Antagonists/pharmacology , Humans , Hydroxymethyl and Formyl Transferases/chemistry , Models, Molecular , Phosphoribosylglycinamide Formyltransferase , Ribonucleotides/chemistry , Structure-Activity Relationship , Tetrahydrofolates/chemistry , Tetrahydrofolates/pharmacology , Tumor Cells, CulturedABSTRACT
The 4-amino-7-oxo-substituted analogues of 5-deaza-5,6,7, 8-tetrahydrofolic acid (5-DATHF) and 5,10-dideaza-5,6,7, 8-tetrahydrofolic acid (DDATHF) were synthesized as potential antifolates. Treatment of the alpha,beta-unsaturated esters 11a-c, obtained in one synthetic step from commercially available para-substituted methyl benzoates (9a-c) and methyl 2-(bromomethyl)acrylate (10), with malononitrile in NaOMe/MeOH afforded the corresponding pyridones 12a-c. Formation of the pyrido[2,3-d]pyrimidines 13a-c was accomplished upon treatment of 12a-c with guanidine in methanol. After the hydrolysis of the ester group present in 13a-c, the resulting carboxylic acids 14a-c were treated with diethyl cyanophosphonate in Et3N/DMF and coupled with L-glutamic acid dimethyl ester to give 15a-c. Finally, the basic hydrolysis of 15a-c yielded the desired 4-amino-7-oxo-substituted analogues 16a-c in 20-27% overall yield. Compounds 16a-c were tested in vitro against CCRF-CEM leukemia cells. The results obtained indicated that our 4-amino-7-oxo analogues are completely devoid of any activity, the IC50 being higher than 20 microg/mL for all cases except 14c for which a value of 6.7 microg/mL was obtained. These results seem to indicate that 16a-c are inactive precisely due to the presence of the carbonyl group in position C7, the distinctive feature of our synthetic methodology.
